[kfaiman]

I just replaced 2x2 EL2001 output buffers w/ 2 EL2009s last night. I'm using them w/ an AD8620 opamp and an Elpac regulated, linear +-12VDC PS. I let it burn in all night... and "burn" it did, they really heat things up. I put a tiny heatsink on each of them (w/ thermal paste), however the Cosmic-sized aluminum enclosure gets rather warm. Maybe I'll have to cut out a vent and add a tiny fan, hehe (brings me back to my overclocking days)...

As for the sound--WOW! The treble is so clear, the way the steel string guitar fades away. The bass is improved as well. I can really feel the air move w/out having to turn it up too high. And the detail is crystal clear. My ATH-W100s now sound better w/ complicated musical passages. I'm very happy with the EL2009s... if only someone can find a large (cheap) batch of them somewhere...

One question, what value resistor are you guys using between the opamp and EL2009s? I have 62 ohms at R7 that I have yet to change, but ppl mentioned that something closer to 150-200 ohm was needed in his setup to reduce ringing. I'll try to see what's actually going on w/ my soon-to-arrive oscilloscope--once I figure out how to use it.

I'll be comparing my META42 to the KG amp/PS this week. Of course the META42 deserves a better PS before I can really compare them fairly.

I'd love to hear from other people's experiences w/ the EL2009s. Thanks.

 

[eric343]

You might be better off thermally coupling the EL2009s to the chassis; the boxes are specifically designed to dissipate heat. But if you decide to go with a fan, make sure it's one that doesn't cause EMI, and it'd probably be easiest to use a really dinky one (like 20-30mm) mounted on the back panel blowing through a 'screen' of small drilled holes and with corresponding holes on the faceplate (maybe get artistic?).

And 'scopes are fairly easy to use... Glad to hear you're getting one, it's what separates an experienced DIYer from a newbie

Check out this guide, it looks handy...: http://www.doctronics.co.uk/scope.htm

Lastly, the "62" ohms is 61.9, as specced by Apheared in his original thread. I believe he did a lot of testing to try and minimize ringing. When picking my resistor values, I tried to stay as close to Apheared's as possible, since he evidently did something right: http://www4.head-fi.org/forums/showt...&threadid=1764

 

[kfaiman]

hi eric,
I wasn't too serious about the fan idea. I'd much rather have the case act as a giant heatsink. I was trying to figure out a way to do it last night, but didn't come up w/ anything. I'm using two different kinds of heatsinks. One is a multi-fin one w/ a flat side that screws to the TO-220 EL2009. The other is one I salvaged from my stripped-down DI/O. It's a slide on type made of one sheet of aluminum that wraps around on itself. It was a tight fit, even had to lay one of the EL2009s flat so the slide-on heatsink had clearance off the board. That Cosmic-case doesn't give you much room to work, hehe. However, the leads are all as short as possible. I'm sure the case would benefit from a vent of some sort.

As for the 61.9 ohm resistor, I'll try out a 150 ohm replacement when I get my oscilloscope and see if it helps. Also, when taking readings, do you use a "test cd" w/ a continuous frequency when measuring the audio signal path? I have no idea how its done.

 

[eric343]

Well, when taking readings, I generally throw in some Pink Floyd, hook it up to my Arcam, and do some heavy duty listening

But if you want to be scientific about it

, then generally either a test CD or dedicated function generator ($9 in kit form from gatewayelex.com) is used, yes. You will need an attenuator of some sort with the kit mentioned above; its output is fairly hot.

 

[puppyslugg]

Hi kfaiman,

I've been using the el2009 in my Corda with the ad8610's or ad843's in the first stage. I've been using a 180 ohm resistor at the input of the el2009. Can't tell if there is ringing or not, no 'scope. But the el2009 sure sounds better than the stock lm6171, and better than the el2008. The el2008 is a very nice buffer, and I would have been happy using the el2008 if it weren't for the el2009. The el2009 seem to have more extension at both ends of the spectrum. I thought it was a better compliment for the ad8610.

The one drawback with the el2009 is: heat. I've been using a clip-on but am looking at getting one those heavier duty types. Will probably drill vent holes in the Corda chassis, too. I want to give the el2009's every chance at having a long existence. They are not cheap and very hard to come by.

 

[ppl]

kfaiman> Glad you liked the EL-2009. as for the 61 ohm resistor that Apheared used. This was for The EL-2001 with the HD-600's connected as a load. To measure Peaking a test CD will not go high enough since Peaking i observed is in the 20 MHz. range. while this may go unnoticed sonicaly to alot of Folks. Critacal listining will reveal smearing of Transients and Hightent sibelence. The Resistor Values i used was for my Layout. If you can confirm that your Amp is stable into your Cans then leave the resistor as is. This resistor value will affect the sound even in stable operation with larger values giving a softer smoother sound and smaller vcalues giving more agressive sonics. IMHO the selection of an Ideal Resistor Value is a subjective matter if the Amp remains stable.

As far as sonicaly the AD-8610 & EL-2009 will sound as good as any Discreet component Amp, However sonic Differences will happen between any two differing Topologies. I am not one that thinks an Opamp based Amp design Properly Done is in any way less better than going Discreet. It is not the Chosen Active devices the designer uses in a Design but rather how the Designer Exploits the advantige and minimizes the drawbacks of a device that matters as far as sound. In addition Opamp based Amp's most often outperform Discreet component designs. Now a well done Discreet component design can perform and sound as good or better than an opamp based one, but the result is more of how well Executed the Design's are.

 

[eric343]

Thanks for the wise words, as always, ppl

I'll have to try out some more values with that middle resistor - the ability to 'tweak' the sound is something I hadn't really thought of as a function of R7.

Live and learn, I guess